Olea Imagein - imaging - Innovation for life - Olea medical
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Olea Imagein
Innovation for life
Issue Number 7- Februar y 2019
ECR Edition
Predicti v e
imaging
■ fMRI ■ COMPUTED MRI ■ STROKE CARE
■ DTI ■ ULTRA-HIGH-FIELD ■ BREAST MRI
■ CEST ■ ELASTOGRAPHY O l■eCARDIAC
a I m a g eMRI
in I n n o v a t i o n for life
#1
Edito
EDITO - Fayçal Djeridane P3
UNDERSTANDING FUNCTIONAL MRI - Interview with Dr. Charles Mellerio P5
PRINCIPLES & APPLICATIONS OF DTI - Interview with Prof. Damien Galanaud P11
MANAGEMENT OF PATIENTS WITH ACUTE STROKE - Dr. Josep Puig, Kambiz Nael & Marco Essig P15
STROKE CARE - Interview with Dr. Noriko Salamon P21
EFFECT OF THE RICIAN NOISE ON THE DETECTION OF ISCHEMIC CORE IN DWI - Timothé Boutelier P25
DEEP DIVE INTO COMPUTED MRI - Interview with Dr. Luca Saba P27
CONTENTS
A NEED FOR ULTRA - HIGH - FIELD MRI - Dr. Makoto Sasaki P33
Fayçal Djeridane
MR IMAGING OF THE ENDOLYMPHATIC HYDROPS IN MENIERE’S DISEASE - Prof. Shinji Naganawa P35
Founder and President of Olea Medical®
APT, GLUCO, IOPAMIDOL - CEST - Interview with Stefano Casagranda P39
CEST CLINICAL APPLICATIONS - Interview with Dr. Sotirios Bisdas P45
Predicting the future is an eternal quest for human kind. A long time ago, shamans were reading
BREAST MRI MANAGEMENT - Interview with Dr. Sophie Taieb P49
the future in their dreams, using their environment and the present to shape the future. As Prof.
ELASTOGRAPHY - Interview with Dr. Denis Le Bihan P53 Elias Zerhouni said “What will exist tomorrow in the medical field already exists today”. The Artificial
MR CARDIAC IMAGING - Interview with Dr. Joao Lima P57 Intelligence is a good example of that. Already in the 50’s, Alan Turing, who can be deemed as
TRADESHOWS & WORKSHOPS P60 one of the fathers of AI, started questioning machine intelligence; but only recently, thanks to the
empowerment of the GPU, it became a reality. It is now feasible to predict Alzheimer’s disease
PLAY TIME P62 6 years in advance!
COMING NEXT P63
The beauty of MR imaging is that we are still far from the full expression of the predictive power for
all the existing or future sequences. The next major steps in MR scan will be standardization and
EDITORIAL BOARD Legal representative:
Fayçal Djeridane
quantitative imaging. For instance, having reliable and repeatable absolute values is mandatory
to diagnose a tumor in an automatic way, predict and follow its response to treatment.
Olea Medical® is a subsidiary
of Canon® Medical Systems Corporation
Printer: Unapei - Entreprise adaptée
les Bambous - 50, avenue Braye de Cau Thirty years ago, when a trauma was arriving in an emergency room, only few images were
13400 Aubagne - France acquired. Now, around 5000 images are produced. Who can read them all in the context of an
Director:
Anca Mitulescu emergency? No one. No one? Not really. AI can do it. Dr Peter Chang, head of the AI laboratory of
Editors: the University of California, Irvine (UCI), has developed an automatic triage of patients in neuro
Brianna Bucciarelli
& Sophie Campana Tremblay emergencies based on CT scan. The first major AI applications in the medical field will be for CT
Graphics: scan. Indeed, the Hounsfield unit standardizes its values.
Gabrielle Croce
Anca Mitulescu, PhD Sophie Campana Brianna Bucciarelli, Reviewer:
Director Tremblay, PhD MSc Margarita Arango For major vendors, healthcare reimbursements are one of the driving forces for the creation
& VP Clinical Affairs Editor Editor Selling price: of research applications. However, hundreds of new start-ups are taking up the challenges
Olea Medical® Clinical & Scientific Clinical Research Free of charge
Research Engineer Engineer Date of legal deposit: left behind by the big companies which are focused on revenue only. They are leading the
Olea Medical® Olea Medical® February 2019 development of artificial pancreas, artificial heart, predictive diagnosis, genomic therapy, etc. In
Publication date: all these challenges MR will play an important role.
February 27th 2019
ISSN Number: 2492-7260
According to the French Data Protection Act of At Olea Medical®, thanks to our astrophysicists, we provided the Bayesian perfusion, which brings
January 6th 1978, modified in 2004,
you have a right of access to and the quantitative perfusion maps to the next level. MR scan is for doctors what telescope is for
rectification of all of your personal data.
astronomers. We are overcoming the limits of knowledge to the utmost patients’ benefit.
If you wish to exercise this right, please send
your request by e-mail to the
Marketing department of the company:
contact@olea-medical.com In this issue, we will present some of the most promising MR techniques.
You can also object, for appropriate Have a good reading!
reasons, to the processing of your
personal data.
Olea Medical® is a French société
anonyme governed by an executive board and
a supervisory board
with a share capital of € 2,040,526.
Registered office:
93 avenue des Sorbiers - ZI Athélia IV
13600 - La Ciotat, France
Ole a Imagein Innovation for life O l ea Im ag e i n Innovation for life
#2 #3
Interview
Understanding
Functional
MRI
“ The playful
aspect of fMRI:
imagine how
to mimic
a daily gesture
in a 70 cm space"
Discover our Charles Mellerio, MD, PhD
SDK
Neuroradiologist at Sainte-Anne Hospital, Paris
and at Centre cardiologique du Nord, Saint-
Denis, France.
Charles Mellerio is a neuroradiologist fo-
cused on two main areas: epilepsy and
Software functional imaging. After completing his
medical training and radiology residency,
Development he specialized in cerebral imaging and
earned his PhD in neurosciences.
Kit His activities as a neuro-radiologist are
conducted over two clinical sites: Sainte-
Anne hospital (Paris) for academic research,
clinical and functional MRI, and Centre Car-
diologique du Nord (Saint-Denis) where fMRI
and advanced imaging are also performed.
O l ea Im ag e i n Innovation for life
Visit olea-medical.com/sdk to learn more #5
Interview
Olea Imagein: Could you briefly overview These sequences result in complex cognitive issues.
the principles of BOLD fMRI based on task For example, if a story listening task is performed
hemodynamic response? versus silence, a cognitive treatment of language
understanding will occur in the brain; but the
Charles Mellerio: Functional MRI records the integration of a noise – i.e. language, will also be
cerebral activity through an indirect process. Why treated. Therefore, two different areas are activated:
indirect? Because, unlike electroencephalography first, the non-language-specific primitive auditory
(EEG) or other techniques, the measure is not linked zone, located in the temporal lobe at the Heschl's
to the electrical activity of the neurons themselves gyrus level and bilaterally activated no matter what
but to their aftereffect, consisting in very local and sound is heard – drill, music or language; second, the
secondary vaso-reactions. When neurons discharge, area we expect to identify and characterize in terms
they need oxygen, provided by hemoglobin, to of laterality, named Wernicke's area and dedicated to
proceed. As a consequence, very local increases – language understanding.
accurate to the nearest millimeter – of cerebral blood
flow occur, with a massive intake of oxyhemoglobin. In other words, the language-specific cognitive
information is drowned into other non-language-
The brilliant point with MRI is that hemoglobin under specific data. Therefore, we have to design pure
its oxygenated or deoxygenated form does not have paradigms able to isolate a particular task, which is
the same signal. It is therefore possible to create brain both interesting and challenging. In the previous
contrast between activated and “rest” areas. example, we can for instance alternate story versus
non-understandable noise listening – what we usually
There are two prerequisite items for fMRI. First, the do is reverse the soundtrack to make it unintelligible
patient must perform a cognitive task in the scanner. while still maintaining a similar acoustic processing.
It can be a simple task, such as moving a hand –
which of course implies a cognitive control of that O.I: What about resting state fMRI?
movement; or more complex tasks, such as exercises
of language, words production, understanding, C.M: We did a summary of the principles of task-based
reading, object recognition or even more high level fMRI, robust and used in clinical routine, but involving
functions dealing with emotions or memory. Of the patient’s cooperation. Besides this method, we
course, the more complex the functions are, the more also have the possibility to conduct resting state
difficult they are to highlight in the brain. Simple or fMRI. This technique does not require the patient’s
primitive functions, i.e. using the primitive motor, contribution, which is a great advantage. The patient
visual or auditory cortex, are easier to isolate. only has to try as much as possible to “switch” his brain
into rest, for a long time – between 5 and 10 minutes.
The second prerequisite item is related to the low He/she is told to let his/her mind wander freely,
orders of magnitude of the measured signals. Indeed, without focusing on a particular idea. The variations
fMRI detects a differential between an activated of the BOLD signal are then recorded, using the same
zone and the remaining resting areas of the brain. signal extraction techniques as in task-based fMRI.
Unfortunately, cerebral rest does not really exist
– except in the deceased subject! Therefore, the What happens is that a brain “at rest” is actually not at
only way to detect a weak signal variation – about all on break. Spatially remote cerebral areas that share
5%, is to repeat the task several times and alternate functional properties, such as sensory-motor, visual
with periods during which the cognitive task is not or executive control networks, are spontaneously
performed. This sequence is called a paradigm. The temporally correlated (i.e. produce a synchronous
most commonly used paradigm in clinical routine signal). Thereby, areas with similar signal evolutions
is designed according to a block strategy, where for over time can be extracted; these areas shape
instance motion is alternated with stillness, or passive networks. Two main methods of analysis are available
text listening is alternated with silence, every 15 to 30 for that purpose: the Independent Component
seconds during 3 to 4 minutes. Analysis (ICA) which blindly extracts all independent
Olea
Ole a Imagein Innovation for life O l ea Im ag e i n Innovation for life
#6 #7
Interview
networks with a statistical approach; and, a region- anatomical landmarks, according to the tumor’s Trans-dural electrodes, positioned near the motor landmarks, the positions of the functional areas are
based technique which starts from defined ROI used location. For example, if the lesion is close to the regions, are indeed indicated for treating neuropathic not reliable anymore, they have to be highlighted by
to calculate correlations with other voxels of the brain. Sylvian fissure either right or left, we will favor pain – for example pain in a leg following a nervous fMRI. Moreover, regarding language, there is a wide
The issue behind this process is that confounding language exercises; more precisely, either language avulsion, or pain related to the phantom limb inter-individual variability; this is precisely where fMRI
connections can be wrongly identified, due to production if the tumor is located in the inferior frontal syndrome. In the latter, we find bias to identify the is interesting; it is not an imaging technique applied
head motion, heart rate or cerebrospinal fluid (CSF) region, or speech comprehension in the temporal former motor functional areas, by asking the patient to a group but to an individual, for a personalized
“
variations for example; a preliminary region. In case of a lesion close to to imagine the movement of the missing limb. medical care.
step is therefore necessary to select the central sulcus, the paradigms
proper networks. The major fMRI will be chosen so as to apply to the The major fMRI applications described above deal O.I: Which developments are still expected
sensorimotor cortex, with motion with neurosurgery, but there are also secondary non- to improve the mapping of cerebral functional
Task-based and taskless fMRI show
complementarity. The most used
applications exercises and tactile stimulation; surgery indications in clinical practice – in research, of areas?
if in the occipital area, the visual course, a wider scope of topics is addressed especially
functional imaging technique
in clinical routine is the task-
described above network will be favored, etc. In in the field of psychiatry. C.M: The analysis of vaso-reactivity, which is the
summary, according to the tumor’s basis of fMRI, can fail if locally modified by a tumor,
based one, because it has been
investigated for a very long time
deal with location, a set of paradigms will be For a very long time before fMRI, clinicians were due to neoangiogenesis process; this is the main bias.
selected, as broad as possible while basing their assumptions on structural anatomical Therefore, improvements could derive from more
”
and has been validated in clinical
studies. In most of the units, resting
neurosurgery still compatible with the machine’s landmarks, on sulcus positions, since we know that accuracy regarding vaso-reactivity, for example using
time – less than 30 minutes; some of them are very stable from one individual a Bayesian method as developed by Olea Medical®.
state is only used in addition to beyond that duration, the patient to another. However, when a tumor distorts those The aim would be to study the hemodynamic
task-based fMRI, in order to provide complementary will have trouble achieving the tasks. During that
information especially for patients whose cooperation period, slightly longer than a conventional MRI, 3 to
is difficult to obtain – children or disabled people. 5 paradigms lasting 3-4 minutes each, in addition to
Moreover, taskless fMRI is a precious research tool for anatomical sequences, are achieved.
pools analysis, e.g. epileptic versus healthy people, in
order to evaluate the networks modifications when The second main indication for fMRI relates to
considering pathologic versus normal condition. patients with an epileptogenic lesion, inducing a
chronic and drug-resistant epilepsy. These patients
O.I: On which criteria do you select the patients may suffer from epileptic seizures several times a
for an fMRI exam? What are the applications day, sometimes since childhood; they are known to
and the associated chosen paradigms? experience a reorganization of the normal functional
areas in the vicinity of the epileptogenic site. If the
C.M: Main indication of fMRI is pre-surgery planning lesion causing epilepsy is for example in the left
for brain tumors. A patient with a brain tumor located temporal lobe, generally associated with language, Left hand Left foot
near a functional area is at high risk of post-operative functional regions of language can move in another
dysfunction if the surgeon damages that critical area, lobe or even on the right side. Therefore, the
linked to a cognitive function, when removing the contralateral hemisphere can be solicited for a task
tumor. fMRI helps mapping the healthy zone near the usually performed in the other side. When the drug-
tumor that needs to be preserved. From a theoretical resistance is diagnosed – in about 50% of the cases,
point of view, it is a very powerful tool. Unfortunately, meaning that no usual medical therapy can treat the
especially for high grade tumors that induce local patient, a focal cortical surgery is proposed to remove
modifications of vaso-reactivity and edema, this the epileptogenic lesion. At that stage, knowledge
can result in false positive and/or false negative of the normal networks together with their potential
responses near the tumor. Therefore, though precious re-organizations is essential. fMRI allows to predict
and systematically performed here in Sainte-Anne and assess the post-operative prognosis regarding
hospital before any awake surgery, task-based fMRI is language, memory, potential dysfunctions, and Left hand
not an exam on which the surgeon can rely 100%. therefore to tailor the surgical procedure. Figure 1: fMRI and DTI in a 45 year-old patient with a right paramedial frontal metastasis using Olea Sphere® software. Anatomical
landmarks predict that this brain tumor is close to the Supplementary motor area (SMA). fMRI is thus performed with 2 motor tasks
Of course, each patient will not perform all the Other indications consist in targeting functional (left foot and left hand) and shows primary motor responses located in the right precentral gyrus, distant to the posterior limits of the
possible paradigms – there are dozens of them. cortical zones with fMRI in order to perform trans- tumor. However, SMA responses are only visualized on the left side and can thus be interpreted as a functional reorganization. The
pyramidal tract is also visualized with DTI close to the posterior part of the lesion. These informations are precious for the neurosurgeon
The choice is established in relation with functional cranial magnetic stimulation or to implant electrodes. and are directly transferred to the operating room.
Ole a Imagein Innovation for life O l ea Im ag e i n Innovation for life
#8 #9
Interview
Principes & Applications
response function (HRF) more individually and more O.I: To which other methods can fMRI be
locally, instead of a canonical approach. Therefore, I combined to capture a larger clinical picture
believe that the main areas for improvement lie in an of the patient?
increased accuracy for HRF estimation, in order to be
of Diffusion Tensor Imaging
as close as possible to the neurons’ electrical activity. C.M: This is a very interesting and important
There is also room for evolution in the design of post- question. fMRI, if considered independently, has no
processing tools, currently separated in two groups: real value. The technique has first to be included
either simplistic, nicely displayed but with poor control within the patient’s anatomy, which requires brain
on the results; either very complex, usually built for segmentation and visualization tools, with optimal
and dedicated to research, requiring programming anatomical sequences. Also, fMRI has to be combined
and unable to provide images that are interpretable and fused with methods able to characterize the "DTI offers two main types of applications:
for a surgeon. The best would be to have a tool at the lesion, such as diffusion and perfusion, in order
interface between clinical and research needs. fMRI to assess its aggressiveness. The problem is that fiber tracking and quantitative imaging"
implies many complex post-processing steps, that currently, too many tools operate independently of
need to be controlled and adjusted if needed. each other; the anatomy, the perfusion and diffusion
data, the tensor imaging information are scattered
The ideal software would allow such a freedom for among different computers or even different rooms.
clinical checking, while remaining user-friendly, fast The radiologist has no choice but to perform a mental
for a proper integration in the clinical environment, gymnastics to mix all the data. Ideally, we would need
robust and accurate. Some of the current tools are powerful and robust synchronization, registration,
easy to use, but their level of confidence is low at the superimposition of these essential findings for a
first sign of trouble. given patient – without falling in the opposite excess:
losing readability because of too many superimposed
Another useful development would be to decrease information.
the duration of the paradigms, the tasks being
sometimes exhausting for the patient. In order to Of course, not to mention that DTI is part of fMRI. DTI
reduce the MR time, the radiologist has to make is mandatory when fMRI is performed, they come
choices, between vision and language for example; together and have to be processed by the same tool.
this can be frustrating. I do not know yet where the This is today fortunately the case, since manufacturers
improvements will come from, maybe from the understood the high complementarity between
paradigm design or from acquisitions themselves (by these two indivisible methods. Damien Galanaud, MD, PhD
acquiring simultaneously multiple slices for example), Professor of Neuroradiology,
but I hope to get the possibility in the future to As a conclusion, I really feel that we are in a transi- La Pitié Salpetrière, Paris, France.
perform more paradigms in less time. tional phase with fMRI; the early marvelous period,
when we were realizing for the first time that we
Also, as radiologists we learnt to refrain from could witness the brain thinking in vivo, is now over. Damien Galanaud is neuroradiologist at
imagining motion not allowed in an MRI scanner, This time was followed by a great disappointment Pitié Salpetrière Hospital. After medicine
such as running; this could be considered as a related to the lack of 100% reliability. Today, we are at studies in Paris and a PhD in Marseille
limitation. However, we can still reproduce many a crossroads, between those who have tremendous directed by Prof. Patrick Cozzone, he is
gestures of everyday life - even playing the piano on expectations and those who would prefer to give now involved in clinical and research
a plastic board, and many different cognitive tasks. up. The coming years will be very open and crucial projects related to head trauma, coma
This is the playful aspect of fMRI: imagine how to to make fMRI become either a powerful and essential and white matter pathology.
mimic a daily gesture in a 70 cm space. tool, or a gadget. I clearly belong to the first group.
His head trauma research is conducted
in association with several international
centers, including a close collaboration
with the Massachussets General Hospi-
tal in the United States.
Ole a Imagein Innovation for life O l ea Im ag e i n Innovation for life
# 10 # 11
Interview
Olea Imagein: Could you shortly summarize O.I: What are the main indications for DTI? same software version can produce slightly different after hypothermia treatment. These indicators
the basics of diffusion tensor imaging (DTI) How relevant is the method for evaluating measures; for that reason, we need extremely stable could be replaced by quantitative diffusion tensor
and tractography techniques? prognoses? sequences and establish processes for normalizing analysis. Indeed, our research group and other teams
the values on controls. demonstrated that DTI could provide an objective
Damien Galanaud: Fundamentally, DTI allows to D.G: From a clinical point of view, open research evaluation of the cerebral damage, and could help
image the white matter structure using the motion is conducted to understand the interconnections Quantitative mapping is used for prognosis in various answering the following question, with more than
properties of water molecules. This technique is based of brain regions. For that purpose, very accurate pathologies, with excellent predictive results in head 95% sensitivity and specificity: will the patient wake
on the primary diffusion imaging sequence, used sequences are required regarding the measure trauma and cardiac arrest. For both pathologies, the up or not? This accurate and reliable evaluation
in stroke diagnosis. However, diffusion does not by of diffusion. This accuracy can be reached if the technique can assess whether the patient will wake up renders the intensive care more secure; if resuscitation
itself inform about the white matter structure, it only spatial resolution is improved and if the parameters and recover, or not. Regarding even mild head trauma, is finally stopped, the clinicians definitely know why
identifies cerebral areas where water diffuses easily or are efficiently computed. To achieve this, the MR studies have shown that DTI could discriminate they made this decision: because the patient had
not – and hence helps detecting an ischemic stroke sequences need to be more and more complex, for between patients with neurologic sequelae, often no chance to survive with acceptable neurological
in the regions with low apparent diffusion coefficient example using Q-ball or NODDI (Neurite Orientation difficult to evaluate, and patients with good recovery. sequels.
(ADC). DTI is a refinement of this technique: it does Dispersion and Density Imaging) models. Getting We can immediately see the potential interest to make
not only evaluate how easily or difficultly water smaller voxels helps better tracking the fibers and an objective assessment of the cerebral injury in head The second type of pathologies is severe head trauma
“
diffusion occurs within the brain, therefore to solve our main issue: trauma patients, with all the implications this may – when people are in coma. As for cardiac arrest,
but also in which direction and with fiber crossing in the brain, meaning have in terms of compensation for the injury. resuscitation is performed a priori; however, the
which properties. In the cerebral
parenchyma, especially within
…tractography that there is an unknown item in
the equation. Advanced sequences Reaching the aim of reproducibility is possible if we
procedure can last for a very long time, without any
clinical or biological tool able to predict the patient’s
white matter, water molecules
are channeled by the axons, the
is extensively used can help assessing the detailed
wiring of the brain and the multiple
scan normal volunteers as controls on the machines
– in that case, it works very well, we have already
outcome, whether he/she will resume a normal life or
will suffer from severe sequelae. Severe sequelae are
dendrites and the myelin sheaths.
Using DTI, we can therefore study
for pre-surgery directions of the fiber tracts. shown that. Of course, we would prefer not to require
these controls by considering several approaches:
defined as people integrating rehabilitation or long-
stay centers, without ever coming back home. Due
both the connections between the
cerebral regions and the integrity of
assessment of As for clinical applications,
tractography is extensively used for
developing phantoms, as for other sequences, but
none really satisfactory are available on the market
to the lack of indicators, clinicians have no choice but
to continue the life support in order to give a chance
”
the axons and the myelin sheaths.
This double ability naturally leads to
cerebral tumors pre-surgery assessment of cerebral
tumors – though it is also sometimes
yet; or using T1 / T2 mapping and myelin imaging to
compensate for the variability.
to any salvageable person; but this is made at a high
cost, by taking the risk of ending up with a vegetative
two main families of applications. performed during a stroke event, in individual and deeply disturbing entire families as has
order to understand if the patient can recover or not. O.I: How predictive and discriminant been publicized by recent famous cases. The high
The first one, fashionable and commonly used in With DTI, the tumor is located with respect to the are the various biomarkers derived from DTI interest of DTI is that it can precisely predict, in 2 out
clinical routine, relates to morphological data regarding white matter tracts, providing information regarding (diffusivity, fractional anisotropy, etc.)? of 3 cases, the patient’s outcome. Whereas cardiac
how the cerebral regions are connected to each other. the surgeon’s margin for tumor excision: is there a arrest algorithm is simple and binary – good or bad
White matter fiber tracking provides this information, risk to cut a track and therefore induce a post-surgery D.G: Let us proceed from the simplest to the most prognosis, the head trauma algorithm we developed
which is interesting in the field of basic research and dysfunction for the patient? This is the first application. complicated case. The simplest is cardiac arrest. We is more subtle and sophisticated; it generates
pre surgical planning of epilepsy and brain tumors. Some work is also conducted on a medullary level, for used to have numerous and very efficient clinical three groups of people: good outcome with 95%
It is usually combined to morphological sequences myelopathy and multiple sclerosis, but this is a much markers for cardiac arrest prognosis. However, these confidence, bad outcome with 95% confidence,
such as 3DT1 or to functional MRI. The second area more unexplored and incidental area of investigation were developed prior to the introduction of recent impossible to categorize the outcome. In two thirds
of applications deals with the quantification of white due to the complexity of implementation; indeed DTI therapies. Indeed, unlike what one might think, of the cases, an answer with 95% confidence is
matter damage, using other parameters linked to resolution is currently insufficient – about 2 mm for resuscitating a patient with cardiac arrest is only the provided, implying a possible use in clinical routine.
water diffusivity; various mappings can be obtained isotropic voxels in routine clinical settings, to image first step of the medical care. In a second time, the In one third of the cases, however, the tool neither
for different white matter diseases. the spinal cord; moreover, bone creates large artefacts person is placed in therapeutic hypothermia condition concludes nor decides – which is better than giving a
on the acquisitions. in order to maintain the brain in a resting state for wrong assumption.
In summary, DTI offers two main types of applications: 48 hours – hypothermia has proven to significantly
fiber tracking for morphological information and For quantitative imaging, reaching a high morpho- improve the patient’s prognosis, even if cardiac arrest The third group relates to subarachnoid hemorrhage.
evaluation of white matter damage using DTI-derived logical accuracy is not as important as for fiber remains a terrible pathology with a 90% mortality For this type of patients, the tool is less efficient and
biomarkers – fractional anisotropy, mean diffusivity, tracking; the essential factor is rather reproducibility. rate. However, our previous clinical biomarkers are no still in development. When an intracranial hematoma
axial diffusivity, radial diffusivity; these markers are Diffusion tensor is extremely sensitive to the variations longer valid with the hypothermia procedure – we is involved, the pathology is even more complex,
able to quantify the white matter status, they hence in acquisition parameters and machines. Two MRI noticed that some patients, unfavorably classified by and research is currently conducted to improve the
belong to the quantitative imaging field. scanners with exactly the same magnet, same coil and these classical biomarkers, were evolving favorably prognosis assessment.
Ole a Imagein Innovation for life O l ea Im ag e i n Innovation for life
# 12 # 13
Article
Management of Patients
with Acute Stroke:
Brain is More Imaging
This work on the prognosis methods [1] is performed This would be a combination of functional and
than Time in the New Era
in collaboration with Prof. Louis Puybasset in the quantitative imaging.
neurosurgical intensive care unit of Pitié Salpétrière
hospital, the engineer Vincent Perlbarg, and Dr Rajiv With MR scanners that are more and more powerful
Gupta from the Massachussets General Hospital in in terms of magnetic fields and gradients, the spatial
Boston. The methods have been developed at Public and anatomical resolution will improve, for a finer
Hospitals of Paris (APHP). To obtain CE-marking and rendering. This will be particularly useful for the mild
FDA-clearance for the software, a startup named head trauma, less easy to analyze than the severe
BrainTale has been created. one. With a higher spatial resolution, an increased
acuteness of quantification and less variability within
O.I: In your opinion, what could be the future the machines, tools will become much more reliable.
potential of tractographic reconstructions
within white matter? The sequences acquired today are so different from
what was achieved 10 years ago; we used to work
D.G: First, I wish that we could obtain quantitative with 27 mm3 voxels in 12 directions, versus 8 mm3 in
models without any necessary control, either by 64 directions today; the size was reduced by a factor
developing phantoms, either by using other cerebral between 3 to 4. I am very confident in all these pieces
markers to correct the variabilities; research is ongoing of technological progress. Regarding the directions,
on the topic. Second, we might consider in the future I do not think it is necessary to increase their numbers
to combine DTI mapping with other quantitative in most applications since studies have shown that
information such as T1, T2, myelin imaging; this would above 30 directions, the models were stable; so, 64
provide finer and more accurate tools. Tractography directions are enough.
and quantification could even be combined, then
they would be able, not only to identify damage in As a conclusion, DTI is a unique sequence able to
anatomical regions, but also to track the fibers and bridge between purely morphological images and
label the regions where the damage is detected. quantification / measurement of brain structures. Josep Puig, MD, PhD
Kambiz Nael, MD
Marco Essig, MD
1. Velly L, Perlbarg V, Boulier T, Adam N, Delphine S, Luyt CE, Battisti V, Torkomian G, Arbelot C, Chabanne R, Jean B, Di Perri C, Laureys S, Citerio G, Vargiolu A, Rohaut B, Bruder N, Girard N, Silva S, Cottenceau V, Tourdias
T, Coulon O, Riou B, Naccache L, Gupta R, Benali H, Galanaud D, Puybasset L, for the MRI-COMA Investigators. Use of Brain Diffusion Tensor Imaging for the Prediction of Long-Term Outcome in Patients after Cardiac
Arrest: a multicentre, prospective, cohort study. The Lancet Neurology. 2018;17(4):317-326.
Ole a Imagein Innovation for life O l ea Im ag e i n Innovation for life
# 14 # 15
Article
A cute ischemic stroke (AIS) is a major cause of
mortality and morbidity worldwide [1,2].
About 15 million people suffer from strokes
each year; of these, 5 million die and another 5 million
end up permanently disabled [2]. Ischemic strokes
far outweigh hemorrhagic strokes, accounting
growing evidence in expanding the therapeutic
window in patients with AIS supports that the use of
advanced imaging techniques to distinguish infarct
core from penumbra is a critical component of the
patient selection process demonstrating the benefit
of mechanical thrombectomy far beyond a 6-hour
the hypoattenuation within affected parenchyma,
the loss of gray-white matter differentiation due to
cytotoxic edema, the ‘insular ribbon sign’, a sulcal
effacement, the ‘hyperdense middle cerebral artery’
(MCA) sign (thrombus within the M1 segment) and/
or the ‘dot sign’ (thrombus within M2) [18]. The recent
likely to benefit from intravenous thrombolysis or
mechanical thrombectomy; the latter should not
be attempted in absence of an identifiable target
on CTA [11,15]. The location of the thrombus
influences the recanalization rates after intravenous
thrombolysis and endovascular therapy [6,7,23,24].
more than 80% of all strokes [2]. In 1996, the Food window [9-15]. To tackle this challenge, the role of guidelines suggest that the presence or absence of Recanalization of an occluded distal internal carotid
and Drug Administration (FDA) approved the use imaging is changing with a remarkable impact on a hyperdense MCA sign should not be used as a artery only occurs in around 5% of patients after
of intravenous (IV) alteplase for the treatment of the diagnostic work-up, treatment decision process criterion for therapeutic decision making purposes. intravenous thrombolysis as opposed to 30% to 90%
AIS within 3 hours of symptom onset. Thus, the and ultimately the treatment itself. With patients for the distal M1 or M2 segments of the MCA [13,24].
National Institutes of Neurological Disorders and now potentially eligible for interventional therapy The most widely used method for quantifying the Large vessel occlusion, longer thrombus, higher
Stroke (NINDS) trial showed that patients with AIS up to 24 hours after the onset of symptoms, the extension of early ischemic changes is the Alberta clot burden and absence of thrombus permeability
treated with intravenous alteplase were 30% more radiologist should expect to see a marked increase Stroke Program Early CT Score (ASPECTS). The MCA are CTA predictors of low recanalization rates after
likely to have good functional outcome at 3 months in imaging requests for stroke, with increased territory is divided into 10 regions, including the intravenous thrombolysis; these patients are more
(defined as a modified Rankin Scale score [mRS] of 0 emphasis on speed and accuracy. In other words, caudate, lentiform nucleus, insula, internal capsule, likely to benefit from additional endovascular
or 1) [3]. Few years later, the European Cooperative as imaging has become the pivotal factor in this and six cortical regions; one point is subtracted for intervention [7,11-15,25]. Therefore, the CTA offers
“
Acute Stroke Study III also demonstrated good process, the term “imaging is brain” would have each region that demonstrates the possibility of promoting
functional outcome when intravenous alteplase become part of the lexicon of stroke diagnosis. imaging findings of acute a more efficient triage of
was administered 3 to 4.5 hours after symptom
onset (52% vs 45%; OR 1.28; 95% CI, 1.0–1.6) [4]. Two facts explain this evolving scenario. With
infarct. Therefore, a score of 10
indicates a normal study, and a
The ultimate goal patients that are candidates
for revascularization therapies.
Recanalization of intracranial thrombus and the
subsequent restoration of blood flow is strongly
the recent publication of the DAWN trial (Clinical
Mismatch in the Triage of Wake-Up and Late
score of zero indicates that the
entire MCA territory is infarcted.
of neuroimaging is A noteworthy fact is that the
2018 AHA/ASA guidelines
associated with improved clinical outcome in
patients with AIS. A meta-analysis encompassing
Presenting Strokes Undergoing Neurointervention
with Trevo) and the DEFUSE-3 trial (Diffusion and
The relationship between
ASPECTS and functional
to help in the triage recommend to study the
extracranial carotid and vertebral
998 patients showed that recanalization significantly
improved 90-day clinical outcome (OR 4.43; 95% CI,
Perfusion Imaging Evaluation for Understanding
Stroke Evolution), the time window to treat has been
outcome after reperfusion is
controversial. Some studies
of patients for arteries in potential candidates
for mechanical thrombectomy,
3.32–5.91) and mortality (OR 0.24; 95% CI, 0.7–17.4)
[5]. However, the location of the clot is one of the
expanded to 24 hours from onset of symptoms
[9,10] on one side, and the 2018 American Heart
found a relationship between
ASPECTS functional outcome
revascularization in addition to intracranial
vessels, in order to provide
”
main determinants of stroke outcome; the ability to
achieve successful recanalization after intravenous
Association/American Stroke Association (AHA/ASA)
guidelines for management of AIS now recommend
after thrombolysis
However, other publications
[19].
therapy useful information on patient
eligibility and endovascular
alteplase administration is limited for large- CT perfusion (CTP), or diffusion-weighted imaging did not [20,21]. More recently, procedural planning. The
vessel occlusion, in particular proximally located (DWI) and perfusion-weighted imaging (PWI) in the the findings from the MR CLEAN trial showed that radiologist should also assess the vasculature of the
clots [6,7]. Recently, Menon et al. demonstrated 6–24-hour time period to facilitate decision making ASPECTS less than 7 did not have a poorer outcome, neck for vessel dissections, stenosis and occlusions
that more distal thrombus location and greater for mechanical thrombectomy on the other [16]. indicating that the extent of early ischemic changes that may assist in planning endovascular procedures
thrombus permeability were associated with vessel The association between endovascular reperfusion at NCCT within the first 6 hours of stroke might or identify which patients are ineligible for treatment
recanalization after administration of intravenous and improved functional outcome is not time not be correlated with functional outcome [15]. because of vessel tortuosity or inability to access the
alteplase; among patients who did not receive dependent in patients with clinical or imaging Although recent guidelines suggest that extension intracranial vasculature [16].
alteplase, rates of arterial recanalization were low [7]. mismatch [17], and that individual patient selection of infarct on NCCT should not be used to decide the
based on imaging can really replace the clock in intravenous thrombolysis, most clinicians prefer to The exclusion of patients who are likely to have
Endovascular therapies are often performed in patients with AIS. know this information when making therapeutic poor outcomes even with prompt revascularization
patients who have received IV alteplase but who decisions, such as mechanical thrombectomy. The is determinant in demonstrating the benefit of
have persistent large vessel occlusion and high Despite the increasing role of more advanced other new recommendation is that multimodal CT mechanical thrombectomy [9-15]. Because DWI is
clot burden [8]. These patients are thought to imaging techniques, the non-contrast CT (NCCT) and MRI, including perfusion imaging, should not superior to CTP not only in detecting the core infarct
respond poorly to IV alteplase. The ultimate goal of is the most commonly used imaging modality delay administration of intravenous alteplase [16]. but also in precisely quantifying infarct volume,
neuroimaging is to help in the triage of patients for for patients with suspected AIS, giving its wider several studies have shown superior outcomes when
revascularization therapy, with the underlying idea availability, fast scanning time, cost-effectiveness The main role of CT angiogram (CTA) is to detect MRI is incorporated into the diagnostic imaging
to select candidates based on individual vascular and and sensitivity to exclude acute hemorrhage [16]. The an intracranial large vessel occlusion, including the work-up of AIS patients [9,10,26,27]. Patients with
physiologic information rather than on rigid time initial role of NCCT is to exclude contraindications to internal carotid artery or M1 segment of the MCA, that a small core infarct are most likely to benefit from
windows. The effectiveness of these therapeutic therapy, such as acute hemorrhage, large infarct or would be amenable to mechanical thrombectomy mechanical thrombectomy [28]. The eligibility for
options is not entirely time dependent. In this line, stroke mimics. The signs of acute ischemia include [22]. Patients with a visible occlusion are more mechanical thrombectomy in the DEFUSE-3 trial
Ole a Imagein Innovation for life O l ea Im ag e i n Innovation for life
# 16 # 17
Article
required a core volume of less than 70 ml on DWI; to the risks of therapy. These patients generally will have been used to infer tissue salvageability and intravenous thrombolysis, whereas patients with
in the DAWN trial, the core ranged from 0 to 50 fare poorly with mechanical thrombectomy even to predict responses to therapy. Various grading poor collaterals did not show a differential effect of
ml, depending on the National Institute of Health with high recanalization rates [31,32]. Conversely, scales have been developed to quantify collateral successful recanalization [39,40]. The ESCAPE trial
Stroke Score (NIHSS) and patient age [9,10]. DWI with a small infarct core and a large penumbra status using CTA or MRA [35]. Multiphase CTA is a used multiphase CTA as a mechanical thrombectomy
excludes patients with large core infarcts and in (i.e. “mismatch”), the risk-benefit analysis would be recently developed technique, quick and easy, to selection tool. Patients with poor collateral status
whom mechanical thrombectomy could result in more favorable for mechanical thrombectomy. assess collateral status [36]. A standard CTA of the were considered a contraindication to mechanical
reperfusion injury, poor functional outcome and Parameters used to define core and penumbra head and neck is obtained in the arterial phase. Two thrombectomy and patients with intermediate
even death [14]. Even when the performance of MRI include mean transit time (MTT), time to maximum additional intracranial scans are obtained in the and good collateral status being supportive of
in the hyperacute setting is logistically feasible, the (Tmax), cerebral blood volume (CBV) and cerebral peak and delayed venous phases. Axial MIP images proceeding to mechanical thrombectomy [41].
imaging work-up of the AIS patient very often begins blood flow (CBF) [8]. There is no clear consensus on are obtained for the arterial, venous and delayed It may be reasonable to incorporate collateral status
“
with NCCT and CTA. If the patient is still considered the specific parameters or thresholds that should phases, and a fast assessment of collateral status of into clinical decision making in some candidates to
as a candidate for mechanical be used to define infarct core good, intermediate and poor can be made. Good determine eligibility for mechanical thrombectomy.
thrombectomy, the patient and penumbra [33,34]. Both collaterals are correlated with decreased infarct core
is sent to MRI for an accurate When risk stratifying the DEFUSE-3 and DAWN trials and penumbra sizes, reduced rate of infarct growth, In summary, the recent trials studying the efficacy
estimation of core infarct. A defined infarct core as relative and improved outcomes [37]. Conversely, poor of mechanical thrombectomy confirm that the
fast (6 min) multimodal MRI patients for mechanical CBF < 30% of normal tissue; collaterals result in decline of ASPECTS likely due association between endovascular reperfusion
protocol with good diagnostic DEFUSE-3 defined penumbra to rapid transformation of ischemic penumbra into and desirable outcomes is not time dependent in
quality has been proposed for thrombectomy, as Tmax > 6 s. In DEFUSE-3, irreversibly infarcted tissue [38]. Collaterals status patients with a perfusion mismatch; thus, individual
the evaluation of patients with CTP criteria for mechanical can be used to select patients for endovascular patient selection, by clinical and imaging criteria,
AIS and, therefore, can result the absolute size of the thrombectomy were infarct therapy. Recent trials have shown how patients with might replace the clock far beyond the 6-hour
in significant reduction in scan core < 70 ml, mismatch volume good collaterals have better functional outcome window. The key question is to accurately identify
time [29]. According to the core and its relative > 15 ml and mismatch ratio ≥ 1.8 after endovascular therapy when compared with patients who are likely to benefit from treatment
recent 2018 AHA/ASA guidelines, [9]. The definition of mismatch
MRI would not necessarily be size to the penumbra on DAWN was more complex:
required in selecting patients infarct core volume less than
”
for mechanical thrombectomy are decisive 21, 31 or 51 ml depending on
Non-contrast
Non-contrast CT
CT at
at admission
admission Non-contrast
Non-contrast CT and CT CT at admission
angiography at admission
in the 0–6 hour time window patient’s age and NIHSS [10].
[16]. However, in the 6–24 hour time window, both DEFUSE-3 used perfusion MRI to randomize patients
the DAWN and DEFUSE-3 trials used CTP, or DWI and with a mismatch profile to endovascular treatment or
PWI, to select patients for mechanical thrombectomy no treatment in the 6- to 16-hour window. Following
[9,10]. The recent guidelines now recommend CTP, or enrollment of approximately 40% of the predicted
CT Perfusion CT Perfusion
DWI and PWI, to be included as part of a standard sample, an interim analysis showed a high likelihood
imaging evaluation for patients within 6–24 hours of benefit in the endovascular group, and the trial
from onset of symptoms to facilitate decision making was terminated. In DAWN, functional outcomes
for mechanical thrombectomy [16]. were better after thrombectomy than with standard
care alone in patients with AIS in the 6- to 24-hour
Infarct core is defined as irreversibly damage tissue window with a mismatch between the severity
DWI at 24 hours DWI at 24 hours
consequently to marked reduction in blood supply. of the clinical deficit and infarct volume assessed
The core is surrounded by the penumbra, a more with CTP or DWI. DAWN and DEFUSE-3 trials have
peripheral region of severely ischemic but potentially been the only randomized clinical trials showing
salvageable tissue [30]. Penumbral tissue is comprised benefit of mechanical thrombectomy more than 6
of stunned cells that have ceased to function properly, hours from the onset of the symptoms. Although
secondary to oligemia. If normal blood supply can be future randomized clinical trials may demonstrate Figure 1: 70-year old man who presented with right hemiparesis Figure 2: 67-year old male presented with right-sided
reestablished through early reperfusion, penumbral that additional eligibility criteria can be used to and aphasia (NIHSS score, 17). On non-contrast CT, early signs of hemiparesis, gaze deviation, as well as aphasia. Non-contrast
tissue can sometimes return to normal function; select patients who could benefit from mechanical ischemia is noted (arrows) with a total ASPECT score of 8. CT head CT shows early and subtle ischemic changes along the
angiography showed occlusion of the left middle cerebral artery left middle cerebral artery territory (blue arrows). CT angiogram
this is the tissue that can potentially be saved by thrombectomy, at this time, the DAWN or DEFUSE-3 M2 segment (not shown). CT perfusion using rCBF< 30% and shows a left MCA M2 segment occlusion (red arrows). Poor
prompt reperfusion. When risk stratifying patients for eligibility should be strictly adhered to in clinical Tmax > 6 seconds in Olea Sphere® software shows an estimated collaterals are seen (arrowheads). CT perfusion using rCBF < 30%
mechanical thrombectomy, the absolute size of the practice [16]. ischemic core of 14 ml and critical hypoperfusion (penumbra) of shows an estimated ischemic core of 112 ml (in red) and no
core and its relative size to the penumbra are decisive. 94 ml shown in red and yellow respectively. Decision was made significant critical hypoperfusion (penumbra) based on Tmax > 6 s
to proceed with endovascular treatment, TICI2B recanalization (in yellow) in Olea Sphere®. Due to presence of large ischemic
With a large infarct core, mechanical thrombectomy Collateral status represents an important factor in was achieved after 3 passes. Follow-up MRI at 24 hours shows core and lack of penumbra, endovascular treatment was not
could yield without benefit and expose the patient the outcome of patients with AIS [8]. Collaterals the final infarction with some areas of petechial hemorrhages. performed. Follow up MRI confirms large established infarction.
Ole a Imagein Innovation for life O l ea Im ag e i n Innovation for life
# 18 # 19Interview
and excluding patients who may be unaffected proportion of patients amenable to treatment. In this
or adversely affected by reperfusion therapies. scenario, the use of advanced imaging techniques
Selection of patients with AIS for revascularization to discriminate infarct core from penumbra is critical.
based on physiologic information may potentially The radiologist must therefore be able to provide
shift the treatment paradigm from a rigid time-based accurate and timely information to assist the clinical
paradigm to a more flexible and individualized, team for appropriate treatment decisions in patients
tissue-based approach, which may increase the with AIS.
1- Department of Radiology, University
of Manitoba, Winnipeg, Canada.
2- Research Unit of Diagnostic Imaging
Institute (IDI), Department of
Radiology [Girona BiomedicalResearch
Stroke
Institute] IDIBGI, Hospital Universitari
Dr Josep Trueta, Girona,Spain.
3- Department of Radiology, Icahn
School of Medicine at Mount Sinai,
Josep Puig1,2, MD, Kambiz Nael3, MD Marco Essig1, MD New York, NY, USA.
Care
PhD
1. Yang Q, Tong X, Schieb L, et al. Vital signs: recent trends in stroke 17. Lansberg MG, Cereda CW, Mlynash M, et al. Diffusion and 30. Heit JJ, Zaharchuk G, Wintermark M. Advanced Neuroimaging
death rates - United States, 2000-2015. MMWR Morb Mortal Wkly Perfusion Imaging Evaluation for Understanding Stroke of Acute Ischemic Stroke: Penumbra and Collateral
Rep 2017;66:933-939. Evolution 2 (DEFUSE 2) Study Investigators Response to Assessment. Neuroimaging Clin N Am. 2018;28:585-597.
2. MacKay J, Mensah GA. Global burden of stroke. In: MacKay J, endovascular reperfusion is not time-dependent in patients 31. Lansberg MG, Straka M, Kemp S, et al. MRI profile and response
Mensah GA. The Atlas of Heart Disease and Stroke. Geneva, with salvageable tissue. Neurology 2015;85:708-714. to endovascular reperfusion after stroke (DEFUSE 2): a
Switzerland: World Health Organization; 2004. 18. Rudkin S, Cerejo R, Tayal A, Goldberg MF. Imaging of acute prospective cohort study. Lancet Neurol. 2012;11:860-7.
3. National Institute of Neurological Disorders and Stroke rt-PA ischemic stroke. Emerg Radiol. 2018 Jul 6. doi: 10.1007/s10140 32. Olivot JM, Mosimann PJ, Labreuche J, et al. Impact of diffusion
Stroke Study Group. Tissue plasminogen activator for acute 018-1623-x. weighted imaging lesion volume on the success of
ischemic stroke. N Engl J Med. 1995;333:1581-1587. 19. Barber PA, Demchuk AM, Zhang J, et al. Validity and reliability endovascular reperfusion therapy. Stroke. 2013
4. Hacke W, Kaste M, Bluhmki E, et al. Thrombolysis with alteplase 3 of a quantitative computed tomography score in predicting Aug;44(8):2205-11.
to 4.5 hours after acute ischemic stroke. N Engl J Med. outcome of hyperacute stroke before thrombolytic therapy. 33. Leiva-Salinas C, Jiang B, Wintermark M. Computed Noriko Salamon, MD
2008;359:1317-1329.
5. Rha JH, Saver JL. The impact of recanalization on ischemic stroke
ASPECTS Study Group. Alberta Stroke Programme Early CT
Score. Lancet 2000 13;355:1670-1674.
Tomography, Computed Tomography Angiography, and
Perfusion Computed Tomography Evaluation of Acute
Professor of Radiology
outcome: a meta-analysis. Stroke 2007;38:967-973. 20. Patel SC, Levine SR, Tilley BC, et al. Lack of clinical significance Ischemic Stroke. Neuroimaging Clin N Am. 2018;28:565-572. and Chief of Neuroradiology,
6. Sillanpaa N, Saarinen JT, Rusanen H, et al. Location of the clot and of early ischemic changes on computed tomography in acute 34. Bristow MS, Simon JE, Brown RAet al. MR perfusion and
outcome of perfusion defects in acute anterior circulation stroke stroke. JAMA 2001;286:2830-2838. diffusion in acute ischemic stroke: human gray and white UCLA, CA, USA
treated with intravenous thrombolysis. AJNR Am J Neuroradiol 21. Dzialowski I, Hill MD, Coutts SB, et al. Extent of early ischemic matter have different thresholds for infarction. J Cereb Blood
2013;34: 100-106. changes on computed tomography (CT) before thrombolysis: Flow Metab. 2005;25:1280-7.
7. Menon BK, Al-Ajlan FS, Najm M, et al. Association of Clinical, prognostic value of the Alberta Stroke Program Early CT Score 35. Raymond SB, Schaefer PW. Imaging Brain Collaterals:
Imaging, and Thrombus Characteristics With Recanalization of in ECASS II. Stroke 2006;37:973-978. Quantification, Scoring, and Potential Significance. Top Magn
Visible Intracranial Occlusion in Patients With Acute Ischemic 22. Smith EE, Kent DM, Bulsara KR, et al. Accuracy of Prediction Reson Imaging. 2017 Apr;26(2):67-75.
Stroke. JAMA 2018;320:1017-1026. Instruments for Diagnosing Large Vessel Occlusion in 36. García-Tornel A, Carvalho V, Boned S, et al. Improving the
8. Lin MP, Liebeskind DS. Imaging of Ischemic Stroke. Continuum Individuals With Suspected Stroke: A Systematic Review for Evaluation of Collateral Circulation by Multiphase Computed
Minneap Minn. 2016;22:1399-1423. the 2018 Guidelines for the Early Management of Patients Tomography Angiography in Acute Stroke Patients Treated
9. Albers GW, Marks MP, Kemp S, et al. Thrombectomy for Stroke With Acute Ischemic Stroke. Stroke. 2018;49:e111-e122. with Endovascular Reperfusion Therapies. Interv Neurol.
at 6 to 16 Hours with Selection by Perfusion Imaging. N Engl J Med.
2018;378:708-718.
10. Nogueira RG, Jadhav AP, Haussen DC, et al.Thrombectomy 6 to
23. Demchuk AM, Goyal M, Yeatts SD, et al. Recanalization and
clinical outcome of occlusion sites at baseline CT angiography
in the Interventional Management of Stroke III trial. Radiology
2016;5:209-217.
37. Piedade GS, Schirmer CM, Goren O, et al. Cerebral collateral
circulation: A review in the context of ischemic stroke and
“Collateral circulation is one Noriko Salamon is Professor of Radiology and
Chief of the Section of Neuroradiology at UCLA
24 Hours after Stroke with a Mismatch between Deficit and
Infarct. N Engl J Med. 2018;378:11-21.
11. Jovin TG, Chamorro A, Cobo E, et al.Thrombectomy within 8 hours
2014;273:202-210.
24. Mishra SM, Dykeman J, Sajobi TT, et al. Early reperfusion rates
with IV tPA are determined by CTA clot characteristics. AJNR
mechanical thrombectomy. World Neurosurg. 2018 Oct 17. pii:
S1878-8750(18)32363-5.
38. Zhang X, Zhang M, Ding W, et al. Distinct predictive role of
of the important factors David Geffen School of Medicine in Los Angeles,
CA, USA. She is a world-renowned neuroradiologist
to determine stroke patient’s
after symptom onset in ischemic stroke. N Engl J Med. Am J Neuroradiol. 2014;35:2265-72. collateral status on clinical outcome in variant stroke subtypes
2015;372:2296-306.
12. Saver JL, Goyal M, Bonafe A, et al. Stent-retriever
25. Tan IY, Demchuk AM, Hopyan J, et al. CT angiography clot
burden score and collateral score: correlation with clinical and
of acute large arterial occlusion. Eur J Neurol. 2018;25:293-300.
39. Berkhemer OA, Jansen IG, Beumer D, et al. Collateral Status on
and a pioneer in epilepsy neuroimaging. For the
last decade, she has specialized in detecting brain
clinical outcome”
thrombectomy after intravenous t-PA vs. t-PA alone in stroke. N radiologic outcomes in acute middle cerebral artery infarct. Baseline Computed Tomographic Angiography and Intra
Engl J Med. 2015;372:2285-2295. AJNR Am J Neuroradiol. 2009;30:525-31. Arterial Treatment Effect in Patients With Proximal Anterior
13. Campbell BC, Mitchell PJ, Kleinig TJ, et. Endovascular therapy for 26. Yoo AJ, Verduzco LA, Schaefer PW, et al. MRI-based selection Circulation Stroke. Stroke. 2016;47:768-76. abnormalities that cause seizures in patients who, in
ischemic stroke with perfusion-imaging selection. N Engl J Med.
2015;372:1009-1018.
for intra-arterial stroke therapy: value of pretreatment
diffusion-weighted imaging lesion volume in selecting
40. Menon BK, Qazi E, Nambiar V, et al. Differential Effect of
Baseline Computed Tomographic Angiography Collaterals
many cases, have normal MRI reports. She developed
14. Bracard S, Ducrocq X, Mas JL, et al. Mechanical thrombectomy patients with acute stroke who will benefit from early on Clinical Outcome in Patients Enrolled in the Interventional the multimodality approach including the process of
after intravenous alteplase versus alteplase alone after stroke recanalization. Stroke. 2009;40:2046-54. Management of Stroke III Trial. Stroke. 2015;46:1239-44.
(THRACE): a randomised controlled trial. Lancet Neurol. 27. Wisco D, Uchino K, Saqqur M, et al. Addition of hyperacute 41. Goyal M, Demchuk AM, Menon BK, Randomized assessment superimposing PET scans that measure the glucose
2016;15:1138-1147. MRI AIDS in patient selection, decreasing the use of of rapid endovascular treatment of ischemic stroke. N Engl J metabolism of the brain onto an MRI scans to
15. Berkhemer OA, Fransen PS, Beumer D, et al. A randomized trial endovascular stroke therapy. Stroke 2014;45:467-72. Med. 2015;372:1019-30.
of intraarterial treatment for acute ischemic stroke. N Engl J Med. 28. Leslie-Mazwi TM, Hirsch JA, Falcone GJ, et al. Endovascular pinpoint the location of a subtle lesion causing the
2015;372:11-20.
16. Powers WJ, Rabinstein AA, Ackerson T, et al. 2018 Guidelines for the
Stroke Treatment Outcomes After Patient Selection Based on
Magnetic Resonance Imaging and Clinical Criteria. JAMA
seizures.
Early Management of Patients With Acute Ischemic Stroke: Neurol. 2016;73:43-9.
A Guideline for Healthcare Professionals From the American 29. Nael K, Khan R, Choudhary G, et al. Six-minute magnetic
Heart Association/American Stroke Association. Stroke resonance imaging protocol for evaluation of acute ischemic
2018;49:e46-e110. stroke: pushing the boundaries. Stroke. 2014;45:1985-91.
Ole a Imagein Innovation for life O l ea Im ag e i n Innovation for life
# 20 # 21You can also read
